U.S. patent number 5,331,427 [Application Number 07/878,335] was granted by the patent office on 1994-07-19 for image reproducing apparatus.
This patent grant is currently assigned to Ricoh Company, Ltd.. Invention is credited to Yoshiyuki Namizuka.
United States Patent |
5,331,427 |
Namizuka |
July 19, 1994 |
Image reproducing apparatus
Abstract
An image reproducing apparatus includes an input part for
inputting digital signals indicating an image and for inputting
supplementary digital signals, a composition part for coding the
digital signals through a prescribed coding process so as to
generate a matrix of blocks in which coded data described by
coefficients produced through the coding process is contained, the
blocks being arranged in two-dimensional coordinates depending on a
level of a frequency corresponding to each of coded digital
signals, a first control part for determining a first region of
low-frequency blocks in which the coded data corresponding to the
digital signals are substantially located within the matrix, and a
second control part for placing allocation data defining a location
of the first region within the matrix into a maximum-frequency
block of the matrix, and for placing data corresponding to the
supplementary digital signals into a second region of
high-frequency blocks of the matrix.
Inventors: |
Namizuka; Yoshiyuki (Yokohama,
JP) |
Assignee: |
Ricoh Company, Ltd. (Tokyo,
JP)
|
Family
ID: |
15068820 |
Appl.
No.: |
07/878,335 |
Filed: |
May 4, 1992 |
Foreign Application Priority Data
|
|
|
|
|
May 8, 1991 [JP] |
|
|
3-131901 |
|
Current U.S.
Class: |
382/239;
358/426.03 |
Current CPC
Class: |
H04N
1/41 (20130101) |
Current International
Class: |
H04N
1/41 (20060101); H04N 001/415 () |
Field of
Search: |
;358/455,433,142,400,426,462,467 |
References Cited
[Referenced By]
U.S. Patent Documents
|
|
|
4237484 |
December 1980 |
Brown et al. |
5051840 |
September 1991 |
Watanabe et al. |
5101438 |
March 1992 |
Kanda et al. |
|
Primary Examiner: Brinich; Stephen
Attorney, Agent or Firm: Mason, Fenwick & Lawrence
Claims
What is claimed is:
1. An image reproducing apparatus for generating a composite image
from (1) image signals indicative of an image and (2) supplementary
signals which are not indicative of the image, said apparatus
comprising:
a) input means for inputting (1) digital signals indicating the
image, and (2) supplementary digital signals which are not
indicative of the image;
b) composition means for coding said digital signals through a
prescribed coding process so as to generate a matrix of blocks
wherein:
1) coded data described by coefficients produced through said
coding process is contained in the matrix of blocks, and
2) said blocks are arranged in two-dimensional coordinates
depending on a level of a frequency corresponding to each of coded
digital signals;
c) first control means for determining a first region of
low-frequency blocks in which said coded data corresponding to said
digital signals is substantially located within said matrix;
and
d) second control means, including:
1) means for placing allocation data defining a location of said
first region within said matrix into a maximum-frequency block of
said matrix, the maximum-frequency block relating to a higher
frequency than the low-frequency blocks, and
2) means for placing data corresponding to said supplementary
digital signals into a second region of high-frequency blocks of
said matrix.
2. An apparatus according to claim 1, wherein said allocation data
and said data corresponding to the supplementary digital signals
are placed in said matrix by said second control means after
adaptive quantization is carried out with respect to each of the
blocks of the matrix.
3. An apparatus according to claim 1, wherein said allocation data
is described with a set of a maximum line value and a maximum
diagonal length value, both said values being determined by said
second control means with respect to the first region of the
low-frequency blocks containing the coded data corresponding to the
digital signals.
4. An apparatus according to claim 1, further comprising first
means for carrying out adaptive quantization for the coded data
corresponding to the digital signals in accordance with a bit rate
being assigned to each of the low-frequency blocks in the first
region, wherein said second control means places the data
corresponding to the supplementary digital signals into the
high-frequency blocks which are different from the low-frequency
blocks containing quantized data corresponding to the digital
signals, and further comprising second means for carrying out
inverse-transform coding of said quantized data supplied by said
first means so as to generate a composite image from the digital
signals and the supplementary digital signals.
5. An apparatus according to claim 1, further comprising a
transform coding part for carrying out orthogonal transform coding
of the digital signals, a run-length coding part for carrying out
run-length coding of the supplementary digital signals, wherein
said first control means comprises a calculating part for
calculating a compression rate required for placing the data
corresponding to the supplementary digital signals within the
matrix of the blocks based on a detected quantity of the data
corresponding to the supplementary digital signals and based on a
detected quantity of the coded data corresponding to the digital
signals.
6. An image reproducing apparatus for separating an image from a
composite image containing (1) image data which is indicative of
the image and (2) supplementary data which is not indicative of the
image, said apparatus comprising:
a) input means for inputting a matrix of blocks corresponding to
said composite image, wherein:
1) said matrix of blocks contains coded data described by
coefficients produced through a prescribed coding process, and
2) said blocks are arranged in two-dimensional coordinates
depending on a level of a frequency of each of coded data
corresponding to said composite image;
b) first transform means for detecting a location of a second
region of low-frequency blocks in which coded data corresponding to
said supplementary data which is not indicative of the image is
located within said matrix by checking allocation data at a
maximum-frequency block of said matrix, the maximum-frequency block
relating to a higher frequency than the low-frequency blocks;
c) separation means for separating said matrix of said blocks, in
accordance with said detected location of said second region,
into:
1) a first region of high-frequency blocks in which coded data
corresponding to image data in said composite image is located,
and
2) said second region in which the coded data corresponding to the
supplementary data is located; and
d) second transform means for carrying out inverse transform coding
of the coded data corresponding to the image data, said coded data
located in said first region of said matrix, so as to output an
image of the resulting image data separately from said composite
image.
7. An apparatus according to claim 6, further comprising means for
decoding the coded data included in the first region supplied by
said separation means so that an original image of the
supplementary data is reproduced.
8. An apparatus according to claim 6, further comprising:
second input means for inputting digital signals indicating an
image and inputting supplementary digital signals;
composition means for coding said digital signals through a
prescribed coding process so as to generate a matrix of blocks in
which coded data described by coefficients produced through said
coding process is contained, said blocks being arranged in
two-dimensional coordinates depending on a level of a frequency
corresponding to each of coded digital signals;
first control means for determining a first region of low-frequency
blocks in which said coded data corresponding to said digital
signals is substantially located within said matrix; and
second control means for placing allocation data defining a
location of said first region within said matrix into a
maximum-frequency block of said matrix, and for placing data
corresponding to said supplementary digital signals into a second
region of high-frequency blocks of said matrix.
9. An apparatus according to claim 8, further comprising switching
means for selecting a contact point between a first point and a
second point, wherein a data composition process is carried out so
as to generate a composite image having supplementary data when the
first point is selected by said switching means, and when the
second point is selected by said switching means a data separation
process is carried out so as to generate an image having no
supplementary data from a composite image.
10. An apparatus according to claim 8, further comprising first
means for carrying out adaptive quantization for the coded data
corresponding to the digital signals in accordance with a bit rate
being assigned to each of the low-frequency blocks in the first
region, wherein said second control means places the data
corresponding to the supplementary digital signals into the
high-frequency blocks which are different from the low-frequency
blocks containing quantized data corresponding to the digital
signals, and further comprising second means for carrying out
inverse-transform coding of said quantized data supplied by said
first means so as to generate a composite image from the digital
signals and the supplementary digital signals.
11. An apparatus according to claim 8, further comprising a
transform coding part for carrying out orthogonal transform coding
of the digital signals, a run-length coding part for carrying out
run-length coding of the supplementary digital signals, wherein
said first control means comprises a calculating part for
calculating a compression rate required for placing the data
corresponding to the supplementary digital signals within the
matrix of the blocks based on a detected quantity of the data
corresponding to the supplementary digital signals and based on a
detected quantity of the coded data corresponding to the digital
signals.
12. An image reproducing apparatus for reproducing a composite
image from (1) halftone image signals which are indicative of a
halftone image and (2) supplementary signals which are not
indicative of the halftone image, said image reproducing apparatus
comprising:
a) input means for inputting (1) digital signals indicating the
halftone image and (2) supplementary digital signals which are not
indicative of the halftone image;
b) switching means, coupled to said input means, for selecting, in
response to an instruction supplied from an operation part, one of
(1) a data composition mode and (2) a data separation mode;
c) transform coding means, coupled to said switching means, for
subjecting said digital signals supplied from said input means to
an orthogonal transform coding process to generate a matrix of
blocks in which coded data described by coefficients produced by
said orthogonal transform coding process is arranged according to a
level of frequency corresponding to each of the coded digital
signals;
d) second coding means, coupled to said switching means, for
subjecting said supplementary digital signals from said input means
to a run-length coding process to generate coded data corresponding
to the supplementary digital signals;
e) data composition means for generating composite image data when
the data composition mode is selected by said switching means, the
generating of the composite image data being based on:
1) the coded data supplied from the transform coding means, and
2) the coded data supplied from said second coding means;
f) means for subjecting said composite image data supplied from
said data composition means to an orthogonal inverse-transform
coding process to reproduce the composite image, and for storing
the composite image in a memory device; and
g) output means for outputting the composite image stored in the
memory device.
13. The apparatus of claim 12, wherein said data composition means
includes:
a) first control means, coupled to said transform coding means, for
placing said coded data corresponding to the digital signals into a
first region of low-frequency blocks within said matrix of
blocks;
b) second control means, coupled to said second coding means, and
including:
1) means for placing said coded data corresponding to the
supplementary digital signals into a second region of
high-frequency blocks within said matrix of blocks, the
high-frequency blocks relating to higher frequencies than the
low-frequency blocks, and
2) means for placing allocation data indicating a location of said
first region within said matrix into a highest-frequency block of
said matrix, the highest-frequency block relating to a higher
frequency than the low-frequency blocks; and
c) means for subjecting said coded data corresponding to said
digital signals to an adaptive quantization process so as to place
quantized data in the low-frequency blocks of the first region of
the matrix; wherein said adaptive quantization process is performed
according to a bit rate assigned to each of the low-frequency
blocks in the first region of the matrix.
14. An image reproducing apparatus for separating an image from a
composite image containing (1) image data describing the image and
(2) supplementary data which does not describe the image, said
apparatus comprising:
a) input means for inputting digital signals indicating the
composite image containing supplementary data;
b) switching means, coupled to said input means, for selecting, in
response to an instruction supplied from an operation part, one of
(1) a data composition mode and (2) a data separation mode;
c) transform coding means, coupled to said switching means, for
subjecting said digital signals supplied from said input means to
an orthogonal transform coding process to generate a matrix of
blocks corresponding to said composite image; wherein coded data
described by coefficients produced by said orthogonal transform
coding process is arranged in two-dimensional coordinates according
to the level of frequency that corresponds to each of the coded
digital signals;
d) data separation means for separating said matrix of blocks
supplied from said transform coding means when the data separation
mode is selected by said switching means, the matrix of blocks
being separated into:
1) a first region of low-frequency blocks in which coded data
relating to said supplementary data is arranged within said matrix,
and
2) a second region of high-frequency blocks in which coded data
relating to halftone image data of said composite image is
arranged, the high-frequency blocks corresponding to higher
frequencies than the low-frequency blocks;
e) means for subjecting said coded data relating to said halftone
image data supplied from said data separation means to an
orthogonal inverse-transform coding process so as to reproduce an
image, and for storing said image in a first memory device;
f) means for subjecting said coded data relating to said
supplementary data supplied from said data separation means to a
decoding process so as to reproduce an original image of the
supplementary data, and for storing said original image in a second
memory device; and
g) output means for outputting said image stored in the first
memory device and said original image stored in the second memory
device.
15. The apparatus of claim 14, wherein said data separation means
includes:
a) means for reading allocation data from said matrix of blocks
supplied by said transform coding means; and
b) means for detecting a location of the first region of
low-frequency blocks within said matrix;
wherein:
1) said allocation data is arranged at a highest-frequency block of
said matrix which relates to higher frequencies than the
low-frequency blocks, and
2) the coded data relating to the supplementary data is arranged in
the first region of low-frequency blocks of said matrix.
16. The apparatus of claim 14, further comprising an image
reproducing apparatus for reproducing a composite image from (1)
halftone image signals which are indicative of a halftone image and
(2) supplementary signals which are not indicative of the halftone
image, said image reproducing apparatus including:
a) input means for inputting (1) digital signals indicating the
halftone image and (2) supplementary digital signals which are not
indicative of the halftone image;
b) transform coding means, coupled to said switching means, for
subjecting said digital signals supplied from said input means to
an orthogonal transform coding process to generate a matrix of
blocks in which coded data described by coefficients produced by
said orthogonal transform coding process is arranged according to a
level of frequency corresponding to each of the coded digital
signals;
c) second coding means, coupled to said switching means, for
subjecting said supplementary digital signals from said input means
to a run-length coding process to generate coded data corresponding
to the supplementary digital signals;
d) data composition means for generating composite image data when
the data composition mode is selected by said switching means, the
generating of the composite image data being based on:
1) the coded data supplied from the transform coding means, and
2) the coded data supplied from said second coding means;
e) means for subjecting said composite image data supplied from
said data composition means to an orthogonal inverse-transform
coding process to reproduce the composite image, and for storing
the composite image in a memory device; and
f) output means for outputting the composite image stored in the
memory device.
17. The apparatus of claim 15, further comprising:
a) first control means, coupled to said transform coding means, for
placing said coded data corresponding to the digital signals into a
first region of low-frequency blocks within said matrix of
blocks;
b) second control means, coupled to said second coding means, and
including:
1) means for placing said coded data corresponding to the
supplementary digital signals into a second region of
high-frequency blocks within said matrix of blocks, the
high-frequency blocks relating to higher frequencies than the
low-frequency blocks, and
2) means for placing allocation data indicating a location of said
first region within said matrix into a highest-frequency block of
said matrix, the highest-frequency block relating to higher
frequencies than the low-frequency blocks; and
c) means for subjecting said coded data corresponding to said
digital signals to an adaptive quantization process so as to place
quantized data in the low-frequency blocks of the first region of
the matrix; wherein said adaptive quantization process is performed
according to a bit rate assigned to each of the low-frequency
blocks in the first region of the matrix.
Description
BACKGROUND OF THE INVENTION
The present invention relates generally to an image reproducing
apparatus, and more particularly to an apparatus for reproducing an
image by coding supplementary digital signals together with image
signals through a transform coding process, the apparatus having
capabilities of inputting and outputting digital signals for
carrying out a text/image composition process and a text/image
separation process.
Conventionally, there is a technique for coding a supplementary
digital signal together with an image signal so as to compress the
range of data to a narrower range, and for decoding the resulting
combined signal. The image signal can be a facsimile signal which
exhibits a high degree of correlation. The digital signal can
represent any other data which is independent of the image signal.
For example, U.S. Pat. No. 4,237,484 discloses such a technique.
There is also a text/image composition technique for combining text
data with a halftone image signal by using a prescribed discrete
orthogonal transform process. In this technique, a redundant
portion of image data produced after the transform process is
performed is replaced by the text data, the resulting image data
containing the text data being encoded into coded information which
have a form of a matrix of blocks in orthogonal transform
coordinates. The redundant portion, which includes frequency
components produced when the transform coding process is performed,
is used for combining the text data with the halftone image data,
and only a small quantity of text data can be combined with the
halftone image data without seriously degrading the picture
quantity.
However, in the above mentioned techniques, no preprocessing of the
supplementary digital signals is performed, and the quantity of the
digital signals to be combined with the image signals is limited
and not reduced to a narrower range so that a large quantity of
supplementary digital signals cannot be combined with the image
signals without degrading the picture quality. Thus, there is a
problem in that it is difficult to reproduce an image containing a
large quantity of text data combined with image signals without
seriously degrading the picture quality.
SUMMARY OF THE INVENTION
Accordingly, it is a general object of the present invention to
provide an improved image reproducing apparatus in which the above
described problems are eliminated.
Another, more specific object of the present invention is to
provide an image reproducing apparatus in which a suitable
preprocessing of supplementary digital signals is carried out and
in which adaptive quantization is carried out for each block of the
matrix of coded picture elements based on an activity level of the
image data, thus reproducing a high-quality composite image with a
large quantity of supplementary information such as text data and
voice data. The above mentioned object of the present invention is
achieved by an image reproducing apparatus which includes an input
part for inputting digital signals indicating an image and for
inputting supplementary digital signals, a composition part for
coding the digital signals through a prescribed coding process so
as to generate a matrix of blocks in which coded data described by
coefficients produced through the coding process is contained, the
blocks being arranged in two-dimensional coordinates depending on a
level of a frequency corresponding to each of coded digital
signals, a first control part for determining a first region of
low-frequency blocks in which the coded data corresponding to the
digital signals is substantially located within the matrix, and a
second control part for placing allocation data defining a location
of the first region within the matrix into a maximum-frequency
block of the matrix, and for placing data corresponding to the
supplementary digital signals into a second region of
high-frequency blocks of the matrix.
Still another object of the present invention is to provide an
image reproducing apparatus which can easily and reliably produce
digital signals indicating text data separated from a composite
image reproduced by the above mentioned apparatus with no need for
considering the existence of a large quantity of text information.
The above mentioned object of the present invention is achieved by
an image reproducing apparatus which includes an input part for
inputting a matrix of blocks corresponding to the composite image,
the matrix of blocks containing coded data described by
coefficients produced through a prescribed coding process, the
blocks being arranged in two-dimensional coordinates depending on a
level of a frequency of each of coded data corresponding to the
composite image, a first transform part for detecting a location of
a second region of low-frequency blocks in which coded data
corresponding to the supplementary data is located within the
matrix by checking allocation data at a maximum-frequency block of
the matrix, a separation part for separating the matrix of the
blocks, in accordance with the detected location of the second
region, into a first region of high-frequency blocks in which coded
data corresponding to image data in the composite image is located
and the second region in which the coded data corresponding to the
supplementary data is located, and a second transform part for
carrying out inverse transform coding of the coded data
corresponding to the image data, the coded data located in the
first region of the matrix, so as to output an image of the
resulting data separately from the composite image.
According to the present invention, it is possible to allow a large
quantity of supplementary digital signals indicating text data or
voice data to be combined with digital signals indicating image
data without seriously degrading the picture quality, thus
reproducing a high-quality composite image. The quantity of the
supplementary data combined with the image data can be varied.
Also, the existence of a large quantity of supplementary digital
signals combined with the digital image signals is not necessary to
be considered, and this feature is useful for reproducing a
composite image of a confidential document.
Other objects and further features of the present invention will
become more apparent from the following detailed description when
read in conjunction with the accompanying drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
FIGS. 1A, 1B and 1C are block diagrams showing embodiments of image
reproducing apparatuses according to the present invention;
FIG. 2 is a block diagram showing the construction of a data
composition part;
FIG. 3 is a flow chart for explaining a text/image composition
process performed by the data composition part;
FIG. 4 is a block diagram showing the construction of a data
separation part;
FIG. 5 is a flow chart for explaining a text/image separation
process performed by the data separation part;
FIGS. 6A and 6B are diagrams for explaining a run-length coding
process performed for text data;
FIGS. 7A and 7B are diagrams for explaining an orthogonal transform
coding process performed for image data;
FIG. 8 is a block diagram showing the construction of a
quantization part;
FIG. 9 is a diagram showing the state of composite images generated
by a data composition part after adaptive quantization is carried
out; and
FIG. 10 is a view showing an example of an operation part in the
image reproducing apparatus.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
A description will now be given of an embodiment of an image
reproducing apparatus according to the present invention. FIG. 1A
shows the construction of this image reproducing apparatus. In FIG.
1A, this apparatus includes an input part 101 for inputting a
digital signal indicative of an image from a document reader or a
supplementary digital signal indicative of text information from a
storage device such as a floppy disk unit, an operation part 102
for inputting an instruction requesting that whether a data
composition process or a data separation process be performed with
the input digital data supplied by the input part 101, a data
composition part 104 for carrying out a text/image data composition
in which the supplementary digital signal is combined with the
image signal, both supplied by the input part 101 through
prescribed coding, a data separation part 105 for carrying out a
text/image data separation in which the added text data is
extracted or separated from the combined image data, and an output
part 107 for outputting such data from either the data composition
part 104 or the data separation part 105 so that the output data is
printed or supplied an external storage device. The output part 105
may include a printer and a storage device as described below.
A switch part 103 switches connection of the input part 101 to
either the data composition part 104 or the data separation part
105, in accordance with the instruction by the operation part 102,
so as to supply the digital signal to either the data composition
part 104 or the data separation part 105. A switch part 106
switches connection of either the data composition part 104 or the
data separation part 105 to the output part 107, so that the data
for which a selected image processing is carried out is supplied to
the output part 107. In the above described image reproducing
apparatus, the data from the input part 101 is selectively supplied
to either the data composition part 104 or the data separation part
105 via the switch part 103 in accordance with the instruction by
the operation part 102. After the data composition process or the
data separation process is carried out for the input signals, the
processed data is supplied to the output part 107 via the switch
part 106. Thus, supplementary digital signals can be combined with
digital image signals through the text/image data composition
process, and the supplementary signals can be extracted from the
combined signals through the text/image data separation process. It
should be noted that the present invention is not limited to the
use of the supplementary digital signals indicating text data, and
that the present invention is also applicable to supplementary
digital signals generated from voice data by carrying out necessary
processes.
FIG. 10 shows an example of the operation part 102 of FIG. 1A. This
operation part is formed with a set of control keys on a control
panel of an image forming system. In this operation part, a switch
key 121 is depressed to change an operating mode between dynamic
mode and high-speed mode, a switch key 122 is depressed to select
an operating process between the data composition process and the
data separation process, and a start key 123 is depressed to start
operation of the image forming system. Also, a clear/stop key is
used when a setting number of copies is changed or a copy repeat
mode is interrupted, and a ten key 125 is used to input a desired
number of copies being reproduced. The function of the operation
part 102 in the image reproducing apparatus of FIG. 1A is achieved
by depressing the switch key 122.
FIG. 1B shows another embodiment of the image reproducing
apparatus, and in this embodiment the data composition part 104 is
selected according to the instruction. In this embodiment, the
input part 101 is connected to the data composition part 104 via
the switch part 103, and the data composition part 104 is connected
to the output part 107 via the switch part 106 so that the above
text/image data composition is achieved. Also, FIG. 1C shows still
another embodiment of the image reproducing apparatus, and in this
embodiment the data separation part 105 is selected according to
the instruction. In this embodiment, the switch part 103 connects
the input part 101 to the data separation part 105, and the switch
part 106 connects the data separation part 105 to the output part
107 so that the above text/image data separation is achieved.
FIG. 2 shows a detailed structure of the data composition part
provided in the image reproducing apparatus of the present
invention. In FIG. 2, the input part 101 includes a reader 201
optically reading a document containing image data or text data and
generating an analog signal indicative of the document data by
photoelectrical conversion of a readout signal, an A/D
(analog-to-digital) converter 202 generating a digital signal
indicative of the document data from the analog signal supplied by
the scanner 201, a storage device 203 storing a digital signal
indicative of image data or text data in a floppy disk or the like,
and a switch part 204 switching connection of the A/D converter 202
and the storage device 203 to the input of the data composition
part 104. A memory 205 stores the digital signal supplied by the
input part 101.
As shown in FIG. 2, the data composition part includes a transform
coding part 206, a memory 207 and a detecting part 208. The
transform coding part 206 carries out orthogonal transform coding
of the digital signal of the image data supplied by the input part
101 so that transform coefficients for each of blocks of the coded
picture elements are generated together with a matrix of blocks of
coded picture elements. The memory 207 stores the transform
coefficients together with the matrix of blocks of the coded
picture elements supplied by the transform coding part 206. The
detecting part 208 detects the quantity of the transform
coefficients for a total of the input picture elements stored in
the memory 207.
The data composition part of the invention basically also includes
a run-length coding part 209 binarizing the digital signal of text
data by a prescribed run-length coding method to generate a
sequence of run-length codes from the digital signal, a memory 210
storing the run-length codes for a total of the input text data,
and a detecting part 211 detecting the quantity of the run-length
codes for a total of the input text data. Based on the detected
quantity of the transform coefficients for a total of the input
image data from the detecting part 208 and based on the detected
quantity of the run-length data for a total of the input text data
from the detecting part 211, a calculating part 212 calculates a
compression rate of the picture elements required for the text data
to be inserted into a remaining vacant region other than the
compressed image data after a prescribed adaptive quantization is
carried out.
In FIG. 2, a quantization part 213 carries out adaptive
quantization for each block in the coded picture elements stored in
the memory 207, in accordance with a suitably assigned bit rate.
The quantized image data supplied by the quantization part 213
forms a compressed image data in which the compression rate
required for the text data to be inserted in the remaining vacant
regions other than the image data is achieved. A composition part
214 carries out a data composition process in which the coded text
data from the detecting part 211 is combined with the compressed
image data from the quantization part 213, thereby transform
coefficients of the combined data. An inverse-transform coding part
215 carries out inverse-transform coding based on the transform
coefficients supplied by the composition part 214, so that a
composite image in which the text data is combined with the image
data is reproduced. A memory 216 stores the composite image
supplied by the inverse-transform coding part 215. Also, in FIG. 2,
the output part 107 includes a printer 217 for printing the data of
the composite image from the memory 216 and a storage device 218
storing and retaining the composite image data from the memory
216.
FIG. 3 shows a text/image composition process performed by the data
composition part of FIG. 2. In a flow chart shown in FIG. 3, step
S301 inputs a digital signal which is supplied from either the
reader 201 or the storage device 203 via the switch part 204, and
step S302 stores the digital signal in the memory 205. Step S303
checks whether the input signal stored in the memory 205 indicates
text data or image data. When it is detected in step S303 that a
digital signal indicating image data is input, a transform coding
process for the image data is carried out by the transform coding
part 206 in step S304. In step S305, the resulting transform
coefficients are stored in the memory 207. Then, in step S306, the
quantity of data of the transform coefficients for a total of the
image data is detected by the detecting part 208.
When it is detected in step S303 that a digital signal indicating
text data is input, step S307 carries out a run-length coding
process for the text data by means of the run-length coding part
209. In step S308, the coded signal is stored in the memory 210.
Then, in step S309, the quantity of the coded text data is detected
by the detecting part 211.
Step S310 detects whether or not both the transform coding process
and the run-length coding process for the two types of information
is completed. Steps S301 through S309 are repeated until both
processes are completed for a total of the input digital signal.
After it is detected in step S310 that both processes are
completed, in step S311, the calculating part 212 calculates a
compression rate of the picture elements required to insert the
text data in the compressed image data, based on the quantity of
the transform coefficients for the total of the image data from the
detecting part 208, and based on the quantity of the coded text
data from the detecting part 211. Also, in step S312, an average
coding rate for the coded picture elements is determined, and a
suitable, different bit rate is assigned to each of the blocks of
the matrix of the coded picture elements, based on the determined
average coding rate and based on the a.c. component of electric
power with respect to transform coefficients of each block.
In step S313, the quantization part 213 carries out adaptive
quantization for each block of the matrix of the coded picture
elements stored in the memory 207 in accordance with the assigned
bit rate so that the data is compressed so as to achieve the
required compression rate. In step S314, the coded text data is
inserted into vacant blocks due to the data compression for
reproducing a composite image. In step S315, allocation data
defining a location of the coded image data is inserted to the
matrix of the coded picture elements at a block representing a
component of maximum frequency due to the adaptive quantization.
Step S316 checks whether or not the steps S313 to S315 are repeated
until the above described process is completed for all the blocks
of the coded picture elements.
After the above described process is completed, in step S317, the
inverse-transform part 215 carries out inverse-transform coding
process based on the transform coefficients supplied by the
composition part 214, so that a composite image is reproduced. In
step S318, the composite image supplied by the inverse-transform
coding part 215 is stored in the memory 216. In step S319, the
stored composite image in the memory 216 is output by the output
part 107.
Accordingly, the preprocessing of text data and the adaptive
quantization based on the transform coefficients or based on the
activity of image data allow a large quantity of text data in
different quantities to be combined with image data without
seriously degrading the picture quality, thus reproducing a
high-quality composite image. The existence of a large quantity of
supplementary digital signals being combined with image signals is
not necessary to be considered, and this feature is useful for
reproducing an image of a confidential document.
Next, a description will be given of the data separation part in
the image reproducing apparatus of the invention. FIG. 4 shows a
detailed structure of the data separation part. In FIG. 4, this
data separation part includes the transform coding part 206, a data
allocation checking part 401, and a separation part 402. The data
allocation checking part 401 checks allocation data in the matrix
of the coded picture elements at a specified block so that the
location of the image data in the matrix is detected in the memory
205. As described above, the allocation data indicates the location
of the image data in the matrix of the composite image and is
already inserted in the coded picture elements at a
maximum-frequency block due to the adaptive quantization. The
separation part 402 carries out a text/image separation process so
that the coded text data is separated from the coded image data
with the transform coefficients in the composite image.
In FIG. 4, the data separation part also includes a decoder 403, a
text-data generating part 404 and a memory 405, in order for
reproducing the text data separately from the image data in the
composite image. The decoder 403 converts run-length signals of the
coded text data into signals indicating levels corresponding to the
run lengths. The text-data generating part 404 generates the text
data from the signals supplied by the decoder 403, the coded text
data being included in the high-frequency blocks of the matrix
indicated by the allocation data. The text data reproduced by the
text-data generating part 404 is stored in the memory 405. The data
separation part also includes a value-zero insertion part 406, an
inverse-transform coding part 407 and a memory 408, in order for
reproducing the image data separately from the text data. The
value-zero inserting part 406 inserts the value zero into vacant
blocks of the composite image data corresponding to the separated
text data as the transform coefficients of the image data supplied
by the data separation part. The inverse-transform coding part 407
carries out inverse-transform coding based on the transform
coefficients of the image data with respect to each of the blocks
of the matrix so that the image data is reproduced. The image data
reproduced by the inverse-transform coding part 407 is stored in
the memory 408 sequentially for each block. A switch part 409
switches connection of the memory 405 storing the text data and the
memory 408 storing the image data to the output part 107.
In FIG. 4, the output part 107 includes a switch part 410, the
printer 217 and the storage part 218. The switch part 410 switches
connection of the data separation part to either the printer 217 or
the storage device 218. Thus, the text data and the image data both
from the data separation part are separately output to either the
printer 217 or the storage device 218.
FIG. 5 shows a text/image separation process performed by the data
separation part of FIG. 4. In a flow chart shown in FIG. 5, step
S501 inputs digital signals indicating a composite image. The
digital signals of the composite image are supplied from the reader
201 and the storage device 203. Each of the digital signals is
stored in the memory 205. Step S502 carries out a transform coding
process for the input digital signal with respect to each block in
the matrix by the transform coding part 206. Step S503 checks
allocation data at a specified block of the matrix by the data
allocation checking part 401, so that the location of the coded
image data in the matrix is detected. Step S504 separates the coded
text data from the remaining image data in the coded picture
elements by means of the separation part 402.
In step S505, the decoder 403 generates signals indicating levels
corresponding to run lengths of the text data from the run-length
signals of the text data supplied by the separation part 402. In
step S506, the text-data generating part 404 reproduces text data
with respect to the block being checked, and the reproduced text
data is stored in the memory 405.
In step S507, the value zero is inserted by the part 406 into each
of blocks in the matrix of the coded picture elements corresponding
to the separated text data, the inserted values being part of the
transform coefficients of the image data supplied by the separation
part 402. In step S508, the inverse-transform coding part 407
carries out inverse-transform coding based on the transform
coefficients of the image data with respect to each of the blocks
of the matrix, so that the image data is reproduced for each block.
In step S509, the image data of each block supplied by the
inverse-transform coding part 407 is stored in the memory 408 for
each block sequentially. The above steps S503 to S509 are repeated
until both processes are completed for all the blocks of the coded
picture elements.
Step S510 checks whether or not both processes of text data and
image data are completed for all the blocks of the coded picture
elements. After the processes are completed, in step S511, the text
data is reproduced by the text-data generating part 404, and the
reproduced text data is stored in the memory 405. In step S512, the
data in the memory 405 is output to the output part 107 via the
switch part 409. Also, in step S513, the image data is reproduced
by the inverse-transform coding part 407, and such data is stored
in the memory 407. In step S514, the data in the memory 408 is
output to the output part 107.
As described above, text data and image data separate from each
other can efficiently be reproduced from a composite image.
Necessary data can safely and easily be obtained from a small
amount of information of the composite image which is compressed at
a high compression rate.
FIGS. 6A and 6B are diagrams for explaining a run-length coding
process which is performed for preprocessing of text data for data
compression. FIG. 6A shows an image of text data obtained by the
reader 201 through raster scanning of a document 601 along scan
lines 602 from side to side. This image of the text data is stored
in the memory 205. FIG. 6B shows a binary signal 603 indicating
coded text data 604 corresponding to a scan line, this signal being
produced by the run-length coding part 209 and the coded text data
being stored in the memory 210 after a run-length coding process is
carried out. The signal 603 of FIG. 6B is generated by raster
scanning the image of the text data along a scan line 602 and by
the A/D conversion. This signal shows a sequence of low-level
portions and high-level portions, the low-level portions each
having a length (Lm0, Lm1, Lm2, . . . ) and the high-level portions
each having a length (Hn0, Hn1, Hn2, . . . ). The coded text data
604 is represented by these signal portions in a form as
"Lm0,Hn0,Lm1,Hn1, . . . ".
FIGS. 7A and 7B are diagrams for explaining an orthogonal transform
coding process which is performed for image data. FIG. 7A shows
image data 701 which is described in two-dimensional coordinates
(on x-axis and y-axis). FIG. 7B shows a matrix of blocks of coded
picture elements which is described by transform coefficients in
frequency-dependent regions. The image data 701 is divided into a
matrix of blocks of "N.times.N" picture elements. For example, a
block 702 is one of the blocks in the image data 701. According to
the present invention, an orthogonal transform coding process is
carried with respect to each of the blocks in the image data, so
that a matrix of blocks of coded picture elements of FIG. 7B
described by transform coefficients along Fx and Fy frequency axes
is produced. In the matrix of FIG. 7B, a left-most top block
represents a transform coefficient corresponding to a d.c.
component of the coded picture elements. High-frequency components
thereof appear in low, right-hand blocks in the matrix. A
right-most bottom block of the matrix represents a transform
coefficient corresponding to a maximum frequency of the coded
picture elements in Fx-Fy coordinates.
FIG. 8 shows the construction of the quantization part 213 in the
image reproducing apparatus. In FIG. 8, a block 801 of coded
picture elements is supplied from the memory 207 to the
quantization part 213, and an adaptive quantization is carried out
for this block, in accordance with a suitably assigned bit rate.
This bit rate 802 is assigned to the block 801, based on a
determined average coding rate and based on the a.c. component of
electric power with respect to transform coefficients of this
block. A selector 803 is provided to select a quantizer 805 from
among a plurality of quantizers SQ1 through SQn having different
quantization steps, in accordance with the assigned bit rate, the
selected quantizer carrying out adaptive quantization for the block
801. A switch part 804 switches connection of the selector 803 to
one of the quantizers SQ1 through SQn. Thus, the quantity of text
data being inserted in the composite image is varied for each of
the blocks of the matrix of the coded picture elements, and the
quantity of data of transform coefficients is optimized so as to
suitably vary the text data quantity.
FIG. 9 shows the state of several composite images generated by the
data composition part after the adaptive quantization is carried
out. In FIG. 9, a matrix of blocks 901 shows a state of a composite
image in which image data and text data almost in the same amount
are combined together. In FIG. 9, blocks 904 indicated by shading
lines and located at low-frequency regions represent transform
coefficients of image data, and blocks 905 indicated by small dots
and located at high-frequency regions represent transform
coefficients of text data. A matrix of blocks 902 indicates a state
of a composite image in which a large quantity of supplementary
text data is combined with a small amount of image data. In this
composite image, the image data blocks usually contain only flat,
simple image data such as background components. A matrix of blocks
903 shows a state of a composite image in which a small amount of
text data is combined with a great amount of text data. In this
composite image, the image data blocks usually contain fine,
complex image data such as outline data and have a large quantity
of high-frequency components. In FIG. 9, a block 906 shows a
maximum-frequency block in the Fx-Fy coordinates in which the
allocation data defining a location of image data in the matrix is
inserted. Since the maximum frequency component of the actual image
data occurs less frequently, a difference is not visually
appreciable to human eye if it is replaced by the allocation data.
According to the invention, the location of image data in the
matrix is defined with a set of a maximum line code Lmax and a
maximum diagonal code Dmax, and this allocation data is denoted as
(Lmax, Dmax). The maximum line code Lmax is a maximum line number
of the coded image data in the matrix and the maximum diagonal code
Dmax is a maximum diagonal line length thereof. For example, the
allocation data for the matrixes of the composite images 901, 902
and 903 shown in FIG. 9 are represented by (5, 4), (3, 2) and (6,
4), respectively. For the sake of convenience, these allocation
data are described with a one-byte hexadecimal code, that is, hex
"54" hex "32" and hex "64". Also, the allocation data is inserted
by the composition part 214 into the maximum-frequency block of the
matrix.
Further, the present invention is not limited to the above
described embodiments, and variations and modifications may be made
without departing from the scope of the present invention.
* * * * *